International Journal of ChemTech...

Microbial production of hydrogen from sorghum stalk

T.R.Manikkandan*

*Bio Process Research Laboratory, Department of Chemical Engineering, Annamalai University, Annamalainagar – 608 002, India.

Abstract : An aerobic strain of α-ProteobacteriaAUChE 103, isolated from sorghum stalk

storage yard has been identified as a potential hydrogen producer. In the present study, the media components and process parameters were optimized for enhanced hydrogen production.

The significant media components namely glucose, malt extract, yeast extract, peptone

andNaCl were determined using Plackett-Burman design. These significant variables were then optimized using central composite design. The optimum conditions were found to be : glucose,

19.25g/L; yeast extract, 3.046g/L; malt extract, 1.64g/L; peptone, 5.640 and NaCl, 4.312g/L.

Box-Behnken design was employed to optimize the process parameters. Under the optimum

conditions a maximum hydrogen yield of 0.91 mol H2/mol glucose was achieved.

1. Introduction

The worldwide energy demand has increased due to rapid growth in population and industrial

development. To meet out the energy demand has become difficult due to fast depletion of fossil fuel reserves. Also combustion of these fossil fuel increases the green house gas emission resulting in global warming and

pollution. Hydrogen is now considered one of the alternatives to fossil fuels [1]. It is preferred to biogas or

methane because hydrogen is not chemically bound to carbon and therefore, combustion does not contribute to

green house gases or acid rainproducing only water [2]. Besides, hydrogen has a high energy yield of 122 KJ/g which is 2.75 times greater than the hydrocarbon fuel [3]. It has been reported that 50 million tons of hydrogen

are traded annually worldwide with a growth rate of nearly 10% per year [4]. Currently, 90% of commercially

usable hydrogen is obtained by steam reformation of natural gas apart from coal gasification and water electrolysis. But these processes are expensiveenergy intensive, require high operating temperatures and are

detrimental to the environment [5, 6].The other methods of hydrogen production are photocatalytic and

biological routes. Hence production of hydrogen by exploiting alternative renewable source seems to gain more

prominence. Biomass and water can be used as renewable resources for hydrogen gas production where biomass is one of the energy sources [7]. Utilization of biomass for hydrogen production through biological

means will be a dual solution for renewable source and less carbon emission process [8]. In the present study,

hydrogen production has been attempted from the renewable lignocellulosic biomass, sorghum stalk using a newly isolated strain of α-ProteobacteriaAUChE 103.

T.R.Manikkandan /International Journal of ChemTech Research, 2019,12(4): 194-209.

DOI= http://dx.doi.org/10.20902/IJCTR.2019.120424

International Journal of ChemTech Research CODEN (USA): IJCRGG, ISSN: 0974-4290, ISSN(Online):2455-9555

Vol.12 No.04, pp 194-209, 2019

http://dx.doi.org/10.20902/IJCTR.2019.120424

T.R.Manikkandan /International Journal of ChemTech Research, 2019,12(4): 194-209. 195

2. Materials and methods

2.1 Isolation of hydrogen producing strain

The strain used in this study was isolated from the soil samples of maize storage yard. Pure culture was

obtained by serial dilution and plating on nutrient agar medium. The following composition: Malt 0.1% (w/v);

yeast extract, 0.2% (w/v); peptone, 0.5% (w/v); NaCl, 0.5% (w/v); glucose, 1.5% (w/v) and agar, 1.5% (w/v) at

pH 7 and temperature of 34oC was used for the growth and agar slants with 1.5% (w/v) of agar was used for the

maintenance of organism.

2.2 Acid hydrolysis

Hydrolysis was carried out by treating 5g of the powdered sorghum stalk with 100ml of 1% H2SO4 for

75 minutes at 121°C and 15psi in an autoclave. After hydrolysis, the hydrolysate was filtered through ordinary filter paper followed by filtration through Whattman No.1 filter paper. The filtrate [63.75 % (v/v) (equivalent to 1.5%

(w/v) glucose)] was mixed thoroughly with the above mentioned media composition and was neutralized with

concentrated NaOH solution to attain neutral pH.

2.3 Batch studies

The pH of the medium was adjusted to 7.0 and was sterilized in an autoclave (121oC and 14 psi) for 20

minutes. The medium was cooled and inoculated with one day pre grown culture [5% (v/v)]. The fermentation was

allowed to take place in a fermentation jar, which is kept in a constant temperature water bath in order to maintain the

fermentation temperatures. The released gas during the fermentation was collected in a separate jar by water displacement method. The gas sample was taken in a syringe and was loaded in to gas chromatography for the

qualitative assay of the hydrogen. All the experimental runs were carried out in triplicate and the average value was

taken.

2.4 Analytical methods

The gas produced during the fermentation was collected in a graduated aspirator bottle by water

displacement method at regular time intervals. The percentage of hydrogen constituted in the total gas was

determined using a gas chromatograph (AIMIL- NUCON 5765, Mumbai, India) equipped with a thermal

conductivity detector and 2.0 m (1/4 in. inside diameter) steel column filled with Porapak Q (50/80 mesh) using nitrogen as carrier gas at a flow rate of 20 ml/min. Injector, oven and column temperature was set at 150°C,

80°C and 200°C respectively.

2.5 Media optimization

The variables which significantly affect the hydrogen yield were screened using Plackett-Burman design. Fifteen variables were screened in 20 experimental runs and the insignificant ones were eliminated. The

statistical software package Minitab version 15.0 was used to analyze the experimental data. Once the critical

factors were identified through screening, the central composite design (CCD) was used to study the effect of

screened variables and their interactive effects on hydrogen production. The effect of significant medium components such as glucose, yeast extract, peptone, malt extract, beef extract and sodium chloride were tested

for their significance on the production of hydrogen.

3. Result and discussion

3.1 Isolation and identification of hydrogen producing strain

Four morphologically different colonies were picked from the agar plates inoculated with serially

diluted soil samples. They were further analyzed for hydrogen production and the strain with the highest

production was chosen for further studies. The nucleotide sequence obtained by the sequencing of PCR amplified 16S rDNA gene was compared with the available database in Genebank. Sequence analysis of 16S

rDNA gene demonstrates that the bacterial isolate is a new clone of α-Proteobacteria and designated as α-

ProteobacteriaAUChE 103.


3.2 Acid hydrolysis

Acid hydrolysis reduces the complexity of the complex substrate and the high yields of sugar after

treatment is necessary for hydrogen fermentability [9, 10]. The batch hydrolysis was carried out for sorghum

stalk in order to obtain an optimum yield of glucose. Sulphuric acid and hydrochloric acid of different concentrations ranges between 0.5% and 5% (v/v) were used in order to find an optimum acid

concentration. All hydrolysis studies were carried out in an autoclave at 121oC and 15psi. Sulphuric acid has

resulted the highest glucose yield (0.34 g glucose/g of substrate) where as the hydrochloric acid resulted a slightly decreased yield (0.30 g glucose/g of substrate). For the case of sulphuric acid the yield was increased steeply when the

acid concentration was increased from 0.5% (v/v) to 1.0% (v/v). But for the case of hydrochloric acid, the yield was

increased continuously up to an acid concentration of 4% (v/v). In both cases the yield was decreased drastically to the

minimum at higher levels of acid concentration. The decrease in yield may be due to the decomposition of glucose at higher level of acid concentrations. A higher acid concentration was also reported to increase the formation of

acetic acid and furfural, which are known for microbial growth inhibition [11]. Among the two acids sulphuric

acid with 1% concentration could be the best choice for the hydrolysis of sorghum stalk and was used throughout the experimentation.

In order to get an optimum time required for the hydrolysis of sugarcane bagasse, the hydrolysis was carried out with 1% (v/v) sulphuric acid with different hydrolysis times varied in the range of 15 to 120

minutes. The maximum glucose yield of 0.39 g/g of sugarcane bagasse was obtained at seventy five minutes

hydrolysis time after which there observed a constant decrease in the yield. The decrease in glucose yield at

higher exposure times may be due to the decomposition of the converted glucose at prolonged hydrolysis. The optimum hydrolysis conditions of 1.0% (v/v) sulphuric acid and seventy five minutes of hydrolysis time were

used for all further experimentation.

3.3 Media optimization

The medium constituents namely, carbon, nitrogen and mineral sources used in the fermentation were screened using a two-level fractional factorial experiment. Fifteen media components were investigated for their

influencein the yield of hydrogen and their coded values and the corresponding actual values are given in

Table3.1. The design was subjected to factorial analysis. Experimental and the predicted values of hydrogen are

presented in Table 3.2.Parity plot between the experimental and the predicted values showed the effects of the variables and their significance on the hydrogen yield (Fig. 1). The P values of the variables are shown in Table

3.3 and those with P values < 0.05 are considered to be significant. The Pareto chart showed that the variables

like, glucose, peptone, yeast extract, malt extract and NaCl have significant effect on hydrogen yield and are selected for further optimization to attain a maximum yield (Fig. 2).

Fig. 1 - Parity plot between the experimental and predicted values of hydrogen yield

(α-ProteobacteriumAUChE 103) from Plackett–Burman design

y = 0.9955x

0.02

0.12

0.22

0.32

0.42

0.52

0.62

0.72

0.02 0.12 0.22 0.32 0.42 0.52 0.62 0.72

H2-Exp (mol H2/mol glucose)

H2-P

red

(m

ol

H2/m

ol

glu

co

se)

Hydrogen-Pred (mol H2/mol

glucose)

Linear (Hydrogen-Pred (mol

H2/mol glucose))


Table 3.1 Variables showing medium components used in Plackett-Burman design for the production of

hydrogen usingα-ProteobacteriumAUChE 103

Variables Low level High level

(–) values

(g/L) (+) values

(g/L) Glucose 20 40

Peptone 1 5

Beef Extract 0.5 2.5

Malt extract 1 3

Yeast extract 2 10

KCl 1 9

NaCl 1 5

NH4Cl 1 9

ZnCl2 0.5 5

KH2PO4 0.05 0.5

K2HPO4 .5 2.5

MnSo4.7H20 0.5 5

MgSo4.7H20 0.1 1

ZnSo4.7H20 0.1 5

FeSo4.7H20 0.1 1

ZnSo4.7H20

NH4Cl

KCl

KH2PO4

MgSo4.7H20

ZnCl2

Beefextract

MnSo4.7H20

K2HPO4

FeSo4.7H20

Yeastextract

NaCl

Peptone

Maltextract

Glucose

876543210

Term

Standardized Effect

2.776

Fig. 2 - Pareto’s Chart for (α-ProteobacteriumAUChE 103) screening of media constituents by Plackett-

Burman design


Table 3.2Twenty run Plackett-Burman design matrix (α-ProteobacteriumAUChE 103) for fifteen variables with the experimental and predicted

H2 yields

Glucose(A); Peptone(B); Beef extract(C); Malt extract(D); Yeast extract(E); KCl(F); NaCl(G); NH4Cl(H); ZnCl2(J); KH2PO4(K); K2HPO4(L);

MnSO4.7H2O(M); MgSO4.7H2O(N); ZnSO4.7H2O(O); and FeSO4.7H2O(P)

Runs A B C D E F G H J K L M N O P

Hydrogen yield

(mol H2/mol

glucose)

Exp. Pred.

1 -1 -1 1 -1 1 -1 1 1 1 1 -1 -1 1 1 -1 0.670 0.7229

2 1 -1 1 1 1 1 -1 -1 1 1 -1 1 1 -1 -1 0.023 0.0170

3 1 -1 1 -1 1 1 1 1 -1 -1 1 1 -1 1 1 0.262 0.2919

4 -1 1 1 -1 -1 -1 -1 1 -1 1 -1 1 1 1 1 0.719 0.6671

5 1 -1 -1 -1 -1 1 -1 1 -1 1 1 1 1 -1 -1 0.232 0.2720

6 -1 1 1 -1 1 1 -1 -1 -1 -1 1 -1 1 -1 1 0.673 0.6560

7 -1 1 -1 1 1 1 1 -1 -1 1 1 -1 1 1 -1 0.638 0.6550

8 -1 -1 -1 1 -1 1 -1 1 1 1 1 -1 -1 1 1 0.432 0.4069

9 1 1 1 1 -1 -1 1 1 -1 1 1 -1 -1 -1 -1 0.673 0.6411

10 1 -1 -1 1 1 -1 1 1 -1 -1 -1 -1 1 -1 1 0.021 0.0150

11 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 0.564 0.5939

12 -1 -1 -1 -1 1 -1 1 -1 1 1 1 1 -1 -1 1 0.696 0.6512

13 1 -1 1 1 -1 -1 -1 -1 1 -1 1 -1 1 1 1 0.139 0.1641

14 1 1 -1 -1 -1 -1 1 -1 1 -1 1 1 1 1 -1 0.720 0.6949

15 -1 1 1 1 1 -1 -1 1 1 -1 1 1 -1 -1 -1 0.531 0.5629

16 1 1 -1 -1 1 1 -1 1 1 -1 -1 -1 -1 1 -1 0.464 0.4192

17 1 1 -1 1 1 -1 -1 -1 -1 1 -1 1 -1 1 1 0.022 0.0759

18 -1 -1 1 1 -1 1 1 -1 -1 -1 -1 1 -1 1 -1 0.494 0.4641

19 1 1 1 -1 -1 1 1 -1 1 1 -1 -1 -1 -1 1 0.696 0.7279

20 -1 1 -1 1 -1 1 1 1 1 -1 -1 1 1 -1 1 0.464 0.5020


Table 3.3 Estimated effects and coefficients of the Plackett–Burman design for α-

ProteobacteriumAUChE 103

Terms Effect Coeffi. SE

Coeffi. T P

Constant

0.4569 0.01827 25.01 0.000

A. -0.2627 -0.1313 0.01827 -7.19 0.002

B. 0.2067 0.1033 0.01827 5.66 0.005

C. 0.0625 0.0312 0.01827 1.71 0.162

D. -0.2257 -0.1129 0.01827 -6.18 0.003

E. -0.1131 -0.0566 0.01827 -3.10 0.036

F. -0.0381 -0.0190 0.01827 -1.04 0.356

G. 0.1535 0.0768 0.01827 4.20 0.014

H. -0.0195 -0.0097 0.01827 -0.53 0.622

J. 0.0533 0.0267 0.01827 1.46 0.218

K. 0.0469 0.0234 0.01827 1.28 0.269

L. 0.0855 0.0428 0.01827 2.34 0.079

M. -0.807 -0.0403 0.01827 -2.21 0.092

N. -0.0533 -0.0267 0.01827 -1.46 0.218

O. -0.0013 -0.0007 0.01827 -0.04 0.973

P. -0.0881 -0.0440 0.01827 -2.41 0.073

Following screening, response surface methodology using CCD was applied to determine the optimal

levels of the significant variables. The coded and the actual values of the five significant medium components used in the design are presented in Table 3.4.The data obtained from the five level central composite design

matrix were used to develop models in which the dependent variable Y (hydrogen yield, mol H2/mol glucose)

was obtained as the sum of the contributions of the independent variables through second order polynomial

equation and interaction terms. The hydrogen yield obtained from experiments and from the model predictions are given in Table 3.5. The correlation coefficient, R

2, between the experimental and predicted data was 0.987,

revealing that 98.7% of experimental data of the hydrogen production was compatible with the data predicted

by the model. The parity plot shows a satisfactory correlation between the experimental and predicted values of the hydrogen production (Fig.3). The experimental results suggest that the minimum and maximum values of

hydrogen yield obtained were0.551mol H2/mol glucose and 0.791mol H2/mol glucose for Run No.50 and Run

No.34 respectively.

Regression analysis of the second order polynomial model for hydrogen production isgiven in Table

3.6. ANOVA indicated that, the linear and quadratic effects of glucose, peptone, malt extract and NaCl, and

interaction effects of glucose-peptone, glucose-malt extract, peptone-sodium chloride and yeast extract-sodium chloride concentration were significant. The production of hydrogen could be predicted by the model:

Y= 0.789 - 0.0038X1 + 0.001X2 + 0.015X3 + 0.0038X4 - 0.009X5 – 0.027X12 – 0.018X2

2

– 0.033X32 – 0.019X4

2 – 0.021X5

2 – 0.018X1X2 + 0.006 X1X3 + 0.012 X1X4 + 0.005X1X5 + 0.014 X2X3 + 0.014

X2X4 – 0.005 X2X5 –0.008X3X4 – 0.006 X3X5 + 0.0068X4X5………. (3.1)

where X1, the glucose concentration (g/L); X2, peptone (g/L); X3, malt extract concentration (g/L); X4, yeast

extract concentration (g/L) and X5, the sodium chloride concentration (g/L).

Hydrogen yield for different levels of variables was predicted from the respective contour plots (Fig. 4

(a-j)). Each contour plot represents an infinite number of combinations of the two test variables with the other

three maintained at their respective zero levels. There was a relative significant interaction between every two variables and the maximum predicted yield was indicated by the surface confined in the smallest ellipse in the

contour plot.


The optimum values obtained by solving the second degree polynomial equation for α-

ProteobacteriumAUChE 103 are glucose, 19.25g/L; peptone, 5.64g/L; malt extract, 1.64g/L; yeast extract, 3.16g/L and NaCl,4.312g/L.

Table 3.4Codes and actual levels of the independent variables for design of experiment for α-

ProteobacteriumAUChE 103 H2 fermentation

Table 3.5Five level factorial central composite design with experimental and predicted values of

hydrogen yield (mol H2/mol glucose) for α-ProteobacteriumAUChE 103 H2 fermentation

Ru

n

No.

Coded values Hydrogen yield (mol H2/mol glucose)

X1 X2 X3 X4 X5 Exp. Pred.

1 1.00 1.00 -1.00 -1.00 1.00 0.571 0.566

2 1.00 1.00 1.00 -1.00 1.00 0.643 0.637

3 0.00 0.00 0.00 2.38 0.00 0.696 0.686

4 0.00 0.00 0.00 0.00 0.00 0.790 0.789

5 1.00 1.00 -1.00 -1.00 -1.00 0.584 0.581

6 1.00 -1.00 1.00 1.00 1.00 0.668 0.683

7 -1.00 1.00 1.00 -1.00 1.00 0.678 0.683

8 0.00 0.00 0.00 0.00 0.00 0.790 0.789

9 -1.00 -1.00 1.00 1.00 -1.00 0.629 0.622

10 -1.00 1.00 1.00 1.00 1.00 0.702 0.694

11 -1.00 -1.00 -1.00 -1.00 -1.00 0.673 0.681

12 0.00 0.00 0.00 0.00 0.00 0.789 0.789

13 -1.00 -1.00 1.00 -1.00 -1.00 0.705 0.699

14 -1.00 1.00 1.00 1.00 -1.00 0.728 0.732

15 0.00 2.38 0.00 0.00 0.00 0.694 0.685

16 1.00 -1.00 1.00 -1.00 -1.00 0.701 0.707

17 0.00 0.00 0.00 -2.38 0.00 0.672 0.668

18 -1.00 -1.00 1.00 1.00 1.00 0.598 0.605

19 1.00 -1.00 -1.00 -1.00 1.00 0.662 0.665

20 1.00 -1.00 -1.00 1.00 -1.00 0.672 0.669

21 0.00 0.00 0.00 0.00 -2.38 0.695 0.687

22 -1.00 1.00 1.00 -1.00 -1.00 0.740 0.749

23 2.38 0.00 0.00 0.00 0.00 0.633 0.623

24 1.00 -1.00 -1.00 -1.00 -1.00 0.652 0.664

25 -1.00 -1.00 1.00 -1.00 1.00 0.643 0.654

26 1.00 -1.00 1.00 1.00 -1.00 0.680 0.679

27 1.00 -1.00 1.00 -1.00 1.00 0.691 0.684

Independent

variables Symbols

Coded levels

-2 -1 0 +1 +2

Glucose, (g/L) X1 10 15 20 25 30 Peptone, (g/L) X2 1 3 5 7 9 Malt extract, (g/L) X3 0.5 1.0 1.5 2.0 2.5 Yeast extract, (g/L) X4 1 2 3 4 5 NaCl, (g/L) X5 1 3 5 7 9


Ru

n

No.

Coded values Hydrogen yield (mol H2/mol glucose)

X1 X2 X3 X4 X5 Exp. Pred.

28 -2.38 0.00 0.00 0.00 0.00 0.644 0.640

29 1.00 1.00 1.00 -1.00 -1.00 0.678 0.682

30 0.00 0.00 0.00 0.00 0.00 0.790 0.789

31 1.00 -1.00 -1.00 1.00 1.00 0.701 0.698

32 -1.00 1.00 -1.00 1.00 -1.00 0.680 0.690

33 -1.00 -1.00 -1.00 1.00 1.00 0.643 0.645

34 0.00 0.00 0.00 0.00 0.00 0.791 0.789

35 1.00 1.00 1.00 1.00 -1.00 0.701 0.714

36 1.00 1.00 -1.00 1.00 1.00 0.645 0.653

37 0.00 0.00 0.00 0.00 0.00 0.790 0.789

38 1.00 1.00 -1.00 1.00 -1.00 0.648 0.646

39 0.00 0.00 0.00 0.00 0.00 0.788 0.789

40 -1.00 -1.00 -1.00 -1.00 1.00 0.674 0.661

41 0.00 0.00 2.38 0.00 0.00 0.643 0.636

42 -1.00 -1.00 -1.00 1.00 -1.00 0.626 0.637

43 -1.00 1.00 -1.00 -1.00 1.00 0.623 0.632

44 0.00 0.00 0.00 0.00 0.00 0.789 0.789

45 0.00 0.00 0.00 0.00 0.00 0.789 0.789

46 1.00 1.00 1.00 1.00 1.00 0.695 0.696

47 0.00 -2.38 0.00 0.00 0.00 0.682 0.678

48 -1.00 1.00 -1.00 -1.00 -1.00 0.685 0.674

49 0.00 0.00 0.00 0.00 2.38 0.648 0.643

50 0.00 0.00 -2.38 0.00 0.00 0.551 0.554

51 -1.00 1.00 -1.00 1.00 1.00 0.676 0.676

52 0.00 0.00 0.00 0.00 0.00 0.790 0.789

Table 3.6Results of the regression analysis of second order polynomial model for theoptimization of

hydrogen production using α-ProteobacteriumAUChE 103

Term

constant Regression

coefficient Std. deviation

T-statistics P-value

Intercept 0.788605 0.002844 277.299 <0.001 X1 -0.003651 0.001375 -2.655 0.012 X2 0.001536 0.001375 1.117 0.272 X3 0.015171 0.001375 11.035 <0.001 X4 0.003857 0.001375 2.805 0.009 X5 -0.009275 0.001375 -6.746 <0.001 X1X1 -0.027689 0.001182 -23.433 <0.001 X2X2 -0.018950 0.001182 -16.038 <0.001 X3X3 -0.033250 0.001182 -28.140 <0.001 X4X4 -0.019656 0.001182 -16.635 <0.001 X5X5 -0.021863 0.001182 -18.503 <0.001 X1X2 -0.018875 0.001600 -11.799 <0.001 X1X3 0.006250 0.001600 3.907 0.001 X1X4 0.012125 0.001600 7.579 <0.001 X1X5 0.005250 0.001600 3.282 0.003 X2X3 0.014438 0.001600 9.025 <0.001 X2X4 0.014938 0.001600 9.337 <0.001


X2X5 -0.005437 0.001600 -3.399 0.002 X3X4 -0.008313 0.001600 -5.196 <0.001 X3X5 -0.006188 0.001600 -3.868 <0.001 X4X5 0.006812 0.001600 4.258 <0.001

Table 3.7Analysis of the variance (ANOVA) for the fitted quadratic polynomial model for the production

of hydrogen using α-ProteobacteriumAUChE 103

Sources of

variation Sum of

squares Degrees of

freedom (DF)

Mean

square

(MS) F- value P-value

Regression 0.191106 20 0.191106 116.68 <0.001 Linear 0.015025 5 0.015025 36.69 <0.001 Square 0.138167 5 0.138167 337.42 <0.001 Interaction 0.037915 10 0.037915 46.30 <0.001 Residual

Error 0.002539 31 0.002539 -

Lack-of-Fit 0.002534 22 0.002534 230.39 <0.001 Pure Error 0.000004 9 0.000004 - - Total 0.193645 51 - - -

Table 3.8Optimum values of the variables obtained from regression equation

Independent variables Optimum value (coded) Optimum value (real) (g/L)

Glucose -0.1683 19.25 Yeast extract 0.0240 3.160 Malt extract 0.3125 1.640 Peptone 0.3125 5.640 NaCl -0.3125 4.312

Fig.3 - Parity plot showing the distribution of experimental versus predicted values of hydrogen yield (α-

ProteobacteriumAUChE 103) by RSM

y = 0.9988x

0.56

0.59

0.62

0.65

0.68

0.71

0.74

0.77

0.56 0.59 0.62 0.65 0.68 0.71 0.74 0.77


H2-P

red

(m

ol

H2/m

ol

glu

co

se)

Hydrogen-Pred (mol H2/mol

glucose)

Linear (Hydrogen-Pred (mol

H2/mol glucose))


(a) (b)

(c) (d)

(e) (f)

-2.37

-1.19

0.00

1.19

2.37

-2.37

-1.19

0.00

1.19

2.37

0.2

0.2475

0.295

0.3425

0.39

H

yd

rog

en

Yie

ld

C: Beef extract E: NaCl -2.37

-1.19

0.00

1.19

2.37

-2.37

-1.19

0.00

1.19

2.37

0.23

0.2675

0.305

0.3425

0.38

H

yd

rog

en

Yie

ld

D: Yeast extract E: NaCl

-2.31

-1.16

0.00

1.15

2.31

-2.31

-1.16

0.00

1.15

2.31

0.43

0.52

0.61

0.7

0.79

H

yd

ro

ge

n Y

ield

A: Glucose B: Peptone

-2.31

-1.16

0.00

1.15

2.31

-2.31

-1.16

0.00

1.15

2.31

0.38

0.485

0.59

0.695

0.8

H

yd

ro

ge

n Y

ield

A: Glucose C: Malt extract

-2.31

-1.16

0.00

1.15

2.31

-2.31

-1.16

0.00

1.15

2.31

0.45

0.535

0.62

0.705

0.79

H

yd

ro

ge

n Y

ield

A: Glucose D: Yeast extract -2.31

-1.16

0.00

1.15

2.31

-2.31

-1.16

0.00

1.15

2.31

0.4

0.5

0.6

0.7

0.8

H

yd

ro

ge

n Y

ield

B: Peptone C: Malt extract


(g) (h)

(i) (j)

Fig. 4 (a-j) - 3D plots showing the interactive effects between significant variables on hydrogen yield in α-

ProteobacteriumAUChE 103 batch fermentation

3.4 Process parameter optimization using Box-Behnken design

The Box-Behnken design was used to optimize the important process parameters namely, initial pH of the

medium, temperature and time of fermentation.The coded and the actual values of the parameters used in the

design are presented in Table 3.9. The design matrix which consists of 15 experimental runs was constructed, in order to arrive at a second order polynomial equation to predict the hydrogen fermentation system. The design

matrix and their corresponding experimental and the predicted values are given in Table 3.10. The highest

hydrogen yield of 0.89 mol H2/mol glucose was obtained for the experimental runs of 1, 4 and 9. The results

were analyzed using the analysis of variance (ANOVA) and the estimated coefficients are presented in Table 3.11. The hydrogen yield using α-ProteobacteriumAUChE 103can be expressed in terms of the following

regression equation;

Y=0.887 + 0.049A + 0.0137B + 0.007C + 3.25AB + 0.015AC + 0.01BC - 0.07A2 -

0.09B2 - 0.07C

2 ……… (3.2)

where, A, pH; B, time and C, the temperature.

The multiple correlation coefficient, R2, obtained from the ANOVA was 0.9998, which indicate that the

model is capable of explaining 99.98% of the variation in response. This is supported by the parity plot between

the experimental and predicted hydrogen yield (Fig. 5).Three dimensional surface plots are drawn to determine

the optimum values and the interactive effect of the three process parameters (Fig. 6 (a-c)). The experimental

-2.31

-1.16

0.00

1.15

2.31

-2.31

-1.16

0.00

1.15

2.31

0.48

0.5575

0.635

0.7125

0.79

H

yd

rog

en

Yie

ld

A: Glucose E: NaCl -2.31

-1.16

0.00

1.15

2.31

-2.31

-1.16

0.00

1.15

2.31

0.52

0.5875

0.655

0.7225

0.79

H

yd

rog

en

Yie

ld

B: Peptone E: NaCl

-2.31

-1.16

0.00

1.15

2.31

-2.31

-1.16

0.00

1.15

2.31

0.44

0.53

0.62

0.71

0.8

H

yd

ro

ge

n Y

ield

C: Malt extract E: NaCl -2.31

-1.16

0.00

1.15

2.31

-2.31

-1.16

0.00

1.15

2.31

0.5

0.5725

0.645

0.7175

0.79

H

yd

ro

ge

n Y

ield

D: Yeast extract E: NaCl


results suggest that the maximum values of hydrogen yield (0.89 mol H2/mol glucose) were obtained for the

runs with the central points.The optimum values obatained by solving the second degree polynomial equation are as follows: pH, 7.0; temperature, 34.5

oC and fermentation time, 42.5h.

Fig. 5 - Parity plot between the experimental and predicted values of hydrogen yield (α-

ProteobacteriumAUChE 103) by Box-Behnken design

Table 3.9The coded and actual values of the variables used in the design

Parameters Coded values

-1 0 1

pH (A) 6.5 7.0 7.5

Time (h) (B) 40 42 44

Temperature (oC) (C) 32 34 36

Table 3.10 Three level Box-Behnken design matrix for the optimization of significant process parameters

in α-Proteobacterium AUChE 103 H2 fermentation

Run No.

pH Temperature Time

Hydrogen yield (mol H2/mol glucose)

Exp. Pred.

1 0.00

0.00

0.00 0.889 0.888

2 -1.00 0.00 -1.00 0.700 0.701

3 -1.00 0.00 1.00 0.685 0.683

4 0.00 0.00 0.00 0.886 0.888

5 1.00 0.00 1.00 0.812 0.812

6 0.00 1.00 1.00 0.747 0.748

7 1.00 1.00 0.00 0.785 0.785

8 0.00 1.00 -1.00 0.715 0.713

y = 1.0002x

0.66

0.71

0.76

0.81

0.86

0.66 0.71 0.76 0.81 0.86


H2-P

red

(m

ol

H2/m

ol

glu

co

se)

Hydrogen-Pred (mol H2/mol glucose)

Linear (Hydrogen-Pred (mol H2/mol

glucose))


9 0.00 0.00 0.00 0.888 0.888

10 -1.00 1.00 0.00 0.680 0.681

11 -1.00 -1.00 0.00 0.660 0.660

12 1.00 0.00 -1.00 0.764 0.766

13 1.00 -1.00 0.00 0.752 0.751

14 0.00 -1.00 1.00 0.698 0.700

15 0.00 -1.00 -1.00 0.707 0.706

Table 3.11 Results of the ANOVA of the process parameter optimization data in α-

ProteobacteriumAUChE 103 H2 fermentation using Box-Behnken design of experiments

Source Sum of Squares

Degrees of

Freedom (DF) Mean Square

F Value

P-value Prob> F

Model 0.12 9 0.013 3331.77 < 0.0001 pH (A) 0.019 1 0.019 4878.74 < 0.0001 Temperature (B)

1.513×10-3 1 1.513×10

-3 392.13 < 0.0001

Time (C) 3.920×10-4 1 3.920×10

-4 101.63 < 0.0001 AB 4.225×10

-5 1 4.225×10-5 10.95 0.0129

AC 9.923×10-4 1 9.923×10

-4 257.25 < 0.0001 B C

4.202×10-4 1 4.202×10

-4 108.95 < 0.0001

A^2 0.022 1 0.022 5678.61 < 0.0001 B^2 0.039 1 0.039 9981.91 < 0.0001 C^2 0.023 1 0.023 6079.10 < 0.0001 Residual 2.700×10

-5 7 3.857×10-6

Lack of fit 1.900×10-5 3 6.333×10

-6 3.17 0.1473 Pure error 8.000×10

-6 4 2.000×10-6

Total 0.12 16

Table 3.12 Estimated coefficient values for α-ProteobacteriumAUChE 103 H2 fermentation

Factor Coefficient Estimate

Degrees of

freedom

(DF)

Standard error

95% CI Low

95% CI High

VIF

Intercept 0.89 1 8.783×10-4 0.88 0.89

pH (A) 0.049 1 6.944×10-4 0.047 0.050 1.00

Temperature

(B) 0.014 1 6.944×10

-4 0.012 0.015 1.00

Time (C) 7.000×10-3 1 6.944×10

-4 5.358×10-3 8.642×10

-3 1.00

AB 3.250×10-3 1 9.820×10

-4 9.280×10-4 5.572×10

-3 1.00

AC 0.016 1 9.820×10-4 0.013 0.018 1.00

BC 0.010 1 9.820×10-4 7.928×10

-3 0.013 1.00

A^2 -0.072 1 9.571×10-4 -0.074 -0.070 1.01

B^2 -0.096 1 9.571×10-4 -0.098 -0.093 1.01

C^2 -0.075 1 9.571×10-4 -0.077 -0.072 1.01

(a) (b)


(c)

Fig. 6 (a-c)- 3D plots showing the interactive effects between the significant process parameters on

hydrogen yield in α-ProteobacteriumAUChE 103 batch fermentation

3.5 Effect of different substrates

Different substrates such as, glucose, sucrose, sugarcane bagasse hydrolysate, sorghum stalk

hydrolysate and cellulose were tested for α-ProteobacteriumAUChE 103 for its ability to utilize variety of

sugars and its hydrogen synthesizing ability. The cheaper lignocellulosic substrate, sorghum stalk hydrolysate

has produced a better hydrogen yield (0.91 mol H2/mol glucose) compared to all synthetic carbon and the other lignocellulosic substrate (sugarcane bagasse hydrolysate) (Fig. 7). In order to understand the kinetics of α-

ProteobacteriumAUChE 103 hydrogen fermentation under different initial substrate concentration, the

experiments were performed at different initial hydrolysate concentrations ranging between 0.5% and 3% (w/v) (glucose equivalent). High initial substrate concentration may play an important role in hydrogen production

rates and yields [12-15]. However, fermenting microorganisms can also have limited tolerance to increased

substrate loading [16-17]. The rate of total gas collection was almost constant throughout the fermentation periods. The percentage hydrogen content in the gas mixture almost doubles during the period of 30 hours to 36

hours. The rate of hydrogen production increases linearly up to 30 hours of fermentation, after that, almost it

doubles up (Fig. 8). The maximum volumetric hydrogen productivity of 8.10 ml/Lh was obtained at a

fermentation period of 42 hours with 1.5% (glucose equivalent) initial substrate concentration.

-1.00

-0.50

0.00

0.50

1.00

-1.00

-0.50

0.00

0.50

1.00

0.66

0.72

0.78

0.84

0.9

H

yd

ro

ge

n Y

ield

A: pH B: Time -1.00

-0.50

0.00

0.50

1.00

-1.00

-0.50

0.00

0.50

1.00

0.68

0.735

0.79

0.845

0.9

H

yd

ro

ge

n Y

ield

A: pH C: Temperature

-1.00

-0.50

0.00

0.50

1.00

-1.00

-0.50

0.00

0.50

1.00

0.69

0.74

0.79

0.84

0.89

H

yd

ro

ge

n Y

ield

B: Time C: Temperature


Fig. 7 - Effect of different substrates on hydrogen yield in α-ProteobacteriumAUChE 103 batch cultures

Fig. 8 - Time profile of hydrogen yield for different substrate concentrations inα-ProteobacteriumAUChE

103 batch culture

4. Conclusions

The study demonstrated the optimization of media components and process parameters for the production of hydrogen using the isolated strain, α-ProteobacteriumAUChE 103.An optimum hydrolysis time

and acid concentration is required for the proper hydrolysis of sorghum stalk, above and below which the

glucose yield was low. A suitable sorghum stalk hydrolysate in the fermentation medium is essential to get higher hydrogen yields; however, an excessively high or low concentration of sorghum stalk hydrolysate will

affects the growth of the organism which will reduce the yield.The optimum concentration of medium

components for α-ProteobacteriumAUChE 103 hydrogen production was found to be: glucose, 19.25g/L; yeast extract, 3.046g/L; malt extract, 1.64g/L; peptone, 5.640 and NaCl, 4.312g/L. Box-Behnken design was used to

test the importance of process parameters on hydrogen production. The optimized values of process parameters

for hydrogen production were as follows: pH, 7.0; temperature, 34.5oC and fermentation time, 42.5h. A highest

hydrogen yield of 0.91 mol H2/mol glucose was achieved, when sorghum stalk hydrolysate (equivalent to 1.5% (w/v)) was used as the carbon source.

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.5 1 1.5 2 3

Concentrations

Hy

dro

gen

yie

ld

(mo

l H

2/m

ol

glu

co

se)

Glucose % (w/v) Sucrose % (w/v)

Cellulose % (w/v) Bagasse hydrolysate % (v/v)Sorghum stalk hydrolysate % (v/v)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 6 12 18 24 30 36 42 48

Time (h)

Hy

dro

gen

yie

ld

(mo

l H

2/m

ol

glu

cose

)

0.5% (v/v)

1 %(v/v)

1.5% (v/v)

2 %(v/v)

3 %(v/v)


References

1. Hsia S, Chou Y. Optimization of Biohydrogen Production with Biomechatronics. J Nanomaterials

2014; 1-11. 2. Nath K, das D. Improvement of fermentative hydrogen production: Various approaches. Appl.

Microbial Biotechnol 2004;65:520-9.

3. Kapdan I K, Kargi F. Bio-hydrogen production from waste material. Enzyme and Microbial Biotechnol

2006;38:569-82. 4. Winter C J. In to the hydrogen economy milestones. Int J Hydrogen Energy 2005;30:681-5.

5. Nath K, Das D. Hydrogen from biomass. Current Science 2003;85.

6. Rollin JA, Martin del Campo J, Myung S, Sun F, You C, Bakovic A, Castro R, Chandrayan SK, Wu CH, Adams MWW, Senger RS, Zhang YHP. High-yield hydrogen production from biomass by in vitro

metabolic engineering: Mixed sugars coutilization and kinetic modeling.Proc. Natl. Acad. Sci2015; 112

(16): 4964–9 7. Saxena R C, Adhikari D K, Goyal H B. Biomass based energy fuel through biochemical routes: A

review. Renewable and Sustainable Energy Reviews 2018;13:167-78.

8. Patrick C H. Fermentative hydrogen production: Principles, progress and prognosis. Int J Hydrogen

Energy 2009; 34: 7379-89. 9. Xing Y, Ma H, Fan Y, Hou H, Chen J.Cellulose-hydrogen production from corn stalk biomass by

anaerobic fermentation.Chinese Sci. Bull.,2009; 54 (8): 1434–41.

10. Nissila ME, Lay C, Puhakka JA. Dark fermentative hydrogen production from lignocellulosic hydrolyzates - A review.Biomass and Bioenergy2014; 67: 145–59.

11. Rai PK, Singh SP, Asthana RK, Singh S. Biohydrogen production from sugarcane bagasse by

integrating dark- and photo-fermentation.Bioresour. Technol2017; 152: 140–6.

12. Kumar N, Das D. Continuous hydrogen production by immobilized Enterobacter cloacae IIT-BT 08 using lignocellulosic materials as solid matrices. Enzyme and Microbial Technol 2001;29:280-7.

13. Lacis LS, Lawford HG. Ethanol production from xylose by Thermoanaerobacterethanolicus in batch

and continuous culture. Archives of Microbiology 1998;150:48-55. 14. Van Ginkel S, Sung S. Biohydrogen production as a function of pH and substrate concentration.

Environ SciTechnol 2001;35:4726-30.

15. Sommer P, Georgieva T, Ahring BK. Potential for using thermophilic anaerobic bacteria for bioethanol production from hemicelluloses. Biochemical Society Transactions 2004;32:283-9.

16. Van Ginkel S, Logan BE. Inhibition of bio-hydrogen production by undissociated acetic and butyric

acids. Environ SciTechnol 2005;39:9351-6.

17. Olsson L, Hahn-Hagerdahl B. Fermentation of lignocellulosichydrolysates for ethanol production. EnzMicrobTechnol 1996;18:312-31.

*****

International Journal of ChemTech...

Documents

Transcript of International Journal of ChemTech...